Chance & Necessity

PhD thesis summaries for the social-media era

It is a bank holiday weekend, and our brains need a rest, and maybe some fun. So try lolmythesis.com. Theses are long, PhD theses are often around 200 pages in physics — hard and time-consuming to write, sometimes quite hard to read. lolmythesis.com is where students boil down their master works into a single, hopefully amusing and illuminating line.

Some of them are just great. The front page features the most recent ones, which changes constantly, but at the moment we have: “Turns out society is screwed without fossil fuels”, which is good to know, and “Sometimes when people don’t say things, they don’t say things differently”, which sounds deep.

Physics degrees can pay off, but will they save us?

Back in 1999, only a couple of years after I was appointed at Surrey, I went to a summer school on statistical physics, in St Andrews, Scotland. One of the lecturers was a guy called Jean-Phillippe Bouchaud. He gave some interesting lectures on the statistical physics of glasses, these are systems like window glass in which the molecules are stuck, at least most of the time. Since then our careers, and I suspect our salaries, have diverged. He is now senior management of France’s largest hedge fund, Capital Fund Management.

Gluing jelly back together

There were a lot of great talks in the Liquids 2014 conference, but one stood out. It stood out because it was based on a beautifully simple idea, and because it might just save lives. The talk was by Ludwik Leibler (ESPCI, Paris). It considered the problem of how to stick two polymer gels together. Polymer gels are soft solids that are mostly water but are solid because long polymer molecules form a network inside them. Many foods are polymer gels. Jelly is a classic example.

Why is evolution so much better than us?

One of the things I work on is the nucleation of crystals. Nucleation is how crystals start to form, and so control over nucleation basically means being able to start a crystal growing where we want and when we want. We scientists have tried really quite hard to do this, for decades, and to be honest we are still pretty rubbish at it. We still can’t predict what we need to do to make a crystal on command. Sometimes nothing happens when we want a crystal to nucleate, other times when we want one crystal to form, instead we get dozens. It is a bit of a tale of woe.

Breaking laws

I have just got back from Liquids 2014, a conference in Lisbon. I thoroughly enjoyed it, and Lisbon was great – sunny, friendly and with great seafood. I will probably write a post or two on the best talks later, but now I want to talk about how I learnt of two scientists trying to reinvent thermodynamics.

Swift movement in shrink-to-fit genes

The remarkable fact that people who work on understanding our DNA always mention in their talks is that we have about 1.5 m of DNA in the nucleus of each of cells, and these nuclei are only around 5 thousandths of a millimetre across. That is a lot of DNA in a small space. Stretched out the DNA is longer than some of us are tall, but this DNA is crammed into a space too small to see with the naked eye.

Monkeying around with HIV

I have just got back from co-organising a science workshop in Lausanne, Switzerland. It was great fun, I thoroughly enjoyed many of the talks. And as an organiser it made me happy to see the scientists enjoying the talks then realising that they can use these ideas in their own work. Some of the attendees who met at the meeting for the first time were even talking of teaming up and working together. If they do, it’ll put a smile on my face that I have helped that.

From the mushy zone to squishy cells

Today I am at conference on metals crystallisation, next
Friday I will be co-chairing a workshop on structure in living cells. These are two of four back-to-back conferences will be attending over the next two weeks. With this many conferences it is good to have variety. This one is near Egham in Surrey, the picture shows part of the venue.

Are less active chromosomes like WD40?

We all have 46 chromosomes worth of DNA in each of our cells. Each one is kind of a stringy object that is typically around a few micrometres across, and all 46 are squeezed into a cell nucleus that is itself maybe only 7 micrometres across. These chromosomes are surprisingly variable. Chromosome number 18 has maybe 250 genes on it, while number 19 has around 1,500 genes, despite being around the same size*. So chromosome 19 is a lot more active than 18, many more proteins are being made from its much larger number of genes.

It has been known for a long time that the chromosomes are not uniformly distributed in the nucleus. The active ones, the ones like chromosome 19 with many genes, tend to be nearer the centre, while the quieter ones are near the edge. You can see this in image and schematic at top left, the less active chromosome 18 is at the bottom near the right end, while the more active chromosome 19 is above it and so away from the edge. The image is a from a paper by Bolzer et al.

Don’t get too excited

What with high pressure penalties, last minute corners, etc, now is a good time to consider the Yerkes-Dodson Law. This is illustrated in the figure to the left. Basically it shows how difficult you find doing a task, e.g., taking a good penalty, is as a function of how much stress you are under/how much adrenalin is in your bloodstream. The y-axis says learning but is known to be more general than just learning.

For simple tasks (solid curve), the more adrenalin, the better, i.e., the more likely it is that you complete the simple task correctly. But for tasks where you have to think (see the dashed curve), it is not so simple. A bit of stress, a bit of adrenalin, helps, but too much hinders you, and the chance of you failing goes back up. There is an optimal level of pressure where you perform your best. Being half asleep is bad, but so is being panicked.